The Abundance-Preference Method for Assemblage Analysis: A Normalized Diagram with Algorithm Implementing the Method

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

General comments

The authors address an issue in the application of the APD method for community ecology. The proposed normalization has the potential to improve the method’s utility. However, the manuscript requires substantial revision to improve clarity, provide stronger methodological justification, present actual results from the case studies comparing normalized ADP and original method, restructure the paper logically, and refine the figures. If the novelty of this manuscript rely solely on log-transformed relative abundances instead of log(abundance) the results need a convincing comparison between the original APD and the normalized APD along with an ecological explanation and significance of the results observed in the case studies. Observe also that transformations of multivariate count data may be problematic in other contexts (Warton, 2018, 10.1111/biom.12728; Warton et al., 2012, 10.1111/j.2041-210X.2011.00127.x); it make sense for the context of ADP? Addressing these points will significantly strengthen the manuscript and clarify the contribution of the normalized APD method.

The mention of an accompanying algorithm with the web application (p. 7, lines 205-206) suggests a practical tool for potential users, increasing the method’s utility if properly documented and accessible. The web application (p. 7, lines 205-206) is promising but lacks detail if it is part of this paper or another (Dakki et al. 2025)? If part of this paper, it needs to be presented (pseudocode or detailed description). If the web application is central, provide the correct, active URL and briefly describe its functionality and input/output formats. Verify and provide the correct URL for the web tool and add a brief description of its use.

Major points:

Introduction

The study currently lacks a clear theoretical foundation linking the concept of ecological niche to that of habitat preference. The ecological niche encompasses a much broader framework than habitats alone and both concepts are not interchangeable (p. 1, lines 31-33). Indeed, the definition of Hutchinson (1957) and that of Soberón & Nakamura (2009, https://doi.org/10.1073/pnas.0901637106) are operationally distinct for a series of applications in ecology. Furthermore, the term “habitat” can carry different meanings depending on context and audience (e.g., Hall et al., 1997, Wildlife Society Bulletin, 25(1):173–182). The manuscript would benefit from a more precise definition of how the authors understand and use the term “habitat” through the paper, as well as an explanation of why focusing on the habitat component of the ecological niche is conceptually and practically important. This clarification has direct implications for the type of data employed (e.g., quantitative habitat measurements or habitat classes) and the methodological approaches used for analysis (e.g., Outlying Mean Index – OMI, Doledec et al., 2000; or the reformulated Indicator Value method by De Cáceres & Legendre, 2009, https://doi.org/10.1890/08-1823.1).

Methods

The manuscript tackles a recognized weakness of the original APD method – its sensitivity to the scale of abundance data and the demographic strategies of the studied species (p. 4, lines 127-137). Normalizing such graphical methods is crucial for broader applicability and comparability in community ecology. It offers a concrete normalization procedure for the abundance axis (y-axis) using a log-ratio transformation (p. 5, Eq. 5) and adapts the zoning hyperbolas to this normalized space (p. 6). However, despite the justification for normalization (p. 4, lines 126-140) is conceptually sound, it could be more compelling. Showing a brief example or simulation illustrating how different abundance scales distort the original APD before introducing the normalization would strengthen the argument.

The paper provides the mathematical formulations for the Degree of Preference (DP), the Exclusive Species Curve (ESC), the normalization, and the hyperbola adjustments (p. 3, 5, 6). The use of nij+1 in the adapted Shannon index (p. 3, Eq. 2) is standard for handling zeros, but a brief explicit mention of this rationale would be helpful for readers less familiar with the original method or information theory applications in ecology. The explanation linking Eq. 3 (p. 3, line 109) to the concept of the Exclusive Species Curve could be elaborated slightly for better conceptual understanding beyond just presenting the formula.

The selection of hyperbola vertices in the original method is termed “experimental” (p. 6, line 181). The rationale for the original values (especially 0.324) and the change to 0.3 in the normalized version needs explanation. The reason for removing the 0.01 hyperbola due to “ecological reasons” (p. 6, line 182) must be explicitly stated. This would make the interpretation clearer. The mathematical description of finding the hyperbola parameter a^2 and the translation vector ui (p. 6, lines 190-202) is dense. Adding a sentence or two explaining the purpose of the optimization problem (finding the vertex on the ESC) and the translation (positioning the standard hyperbola) in simpler terms could improve accessibility. Why is the lowest preference class (accidental/foreign) less relevant or problematic at DP < 0.1? Why the authors didn’t provide the preference classes in figure 3 (p. 6, line 185)?

Results

The current “Results” section (Section 3) is confusingly structured. It starts by reminding the reader of the original APD method (3.1), then discusses the normalization (3.2), and finally mentions the algorithm (3.3). Sections 3.1 and 3.2 feel more like extended Methods or Introduction components.

A dedicated “Methods” section should clearly describe: (a) the original APD method briefly, (b) the rationale and procedure for normalization, (c) the adaptation of the zoning, and (d) the case studies used (I couldn't find the supplementary materials!).

A new dedicated “Results” section should present the outcomes of applying the normalized method to the case studies, including new figures. Move the detailed description of the normalization to the Methods section. Populate the Results section with findings from the case studies.

The figures require significant improvement. All figures need clear axis labels with units or descriptions:

Figure 1: I recommend use “Abundance (log2(n))” for the y-axis, “Degree of Preference (DP)” for the x-axis. The legend also needs to be clearer; explain what the different zones (Foreign, Tolerant, etc.) represent ecologically in the methods. The source “(Dakki 1987)” should be the proper citation format for readers reach it in the reference list.

Figure 2 (p. 5): Needs clearer labels. Observe that the y-axis  here is “Degree of Preference (DP)”. It makes sense for me to standardize the name of the axes along all figures.

The caption mentions a “geometric-progression of 1.1-step” – briefly explain how this simulation was set up and why it’s relevant for illustrating the point about n+1 transformation.

Figure 3 (p. 6): I suggest change the y-axis to Normalized Abundance once defining its the conceptual formulation in the methods. The legend needs to clearly map each curve style/color to the number of habitats (J). The caption should reiterate that this shows the ESC shape variation with J after normalization.

Redraw all figures with standard, clear labels and legends. Include figures showing the application of the normalized APD to the case study data, ideally comparing it to the original APD for the same data to demonstrate the benefits of normalization. The Results section needs to present these findings, not just describe the method again.

Discussion

This section should interpret the results, highlight the advantages and limitations of the normalized method compared to the original, discuss the implications for community analysis, and suggest future directions.

The manuscript mentions using case studies (insects, waterbirds) (p. 2, lines 72-76) to illustrate the adaptations. The intention to illustrate the method using real datasets (running water insects and waterbirds, p. 2, lines 72-76) is a strength, although the actual results from these datasets are not compared with the original method in the current draft.

No results (e.g., actual APDs from these datasets comparing normalized vs. non-normalized versions) are really presented. This is a major omission. The results section (Section 3) primarily re-explains the method rather than presenting findings from its application.

Minor Points:

p. 1, line 13: “provide” should be “provides”.

p. 1, line 14: “of an assemblages” should be “of an assemblage” or “of assemblages”.

p. 1, line 16: “analyzes” should be “analyses”.

p. 1, line 17: “showed also” change to “also showed”.

p. 1, line 23: “does not affect” change to “do not affect”.

p. 2, line 43: “aim at quantifying” change to “aim to quantify”.

p. 2, line 54: “need of normalizing” use “need for normalizing”.

p. 5, line 161: “As it is recalled above” change to “As recalled above”.

Comments on the Quality of English Language

The manuscript’s clarity needs improvement. Several sentences are overly long making the text difficult to follow. The language should be more direct and concise throughout the manuscript. Undertake a thorough revision of the text for clarity, conciseness, and grammatical accuracy. Break down long sentences. Ensure smooth transitions between paragraphs and sections. For instance, the abstract (p. 1, lines 12-24) could be rewritten for better flow and immediate clarity on the core problem and proposed solution. The manuscript would also benefit from overall English editing to improve clarity and conciseness.

Author Response

We would like sincerely thank reviewers for their thoughtful and constructive feedback on our manuscript. Their insightful comments have been instrumental in improving the clarity, structure, and scientific rigor of our work. We have carefully revised the manuscript in accordance with their suggestions, including a more precise theoretical positioning, clearer methodological explanations, and the inclusion of applied results on real datasets.

Here are some key answers to your suggestions and questions

 

Reviewer 1

Comment 1: The conceptual link between ecological niche and habitat preference, clarifying the distinction between these two notions, and providing a clear definition of the term "habitat".

Response: In this study, the term 'habitat' refers to the homogeneous landscape units. The niche concept has been removed from the text.

Comment 2: The suitability of the APD approach and the methods used, particularly in comparison with other existing multivariate methods (OMI, IndVal).

Response: As discussed in Dakki et al. (2025), we acknowledge that the APD method can be complemented by other existing multivariate approaches, such as OMI (Dolédec et al., 2000) and IndVal (De Cáceres & Legendre, 2009). However, the aim of this research is not to compare these methods. [Page 2, lines 58–61]

Comment 3: The use of nij + 1 in the adapted Shannon index.

Response: This clarification is addressed on [Page 2, lines 75–76] and has been explained by Dakki et al 2025).

Comment 4: The explanation linking Eq. 3 (p. 3, line 109) to the concept of the Exclusive Species Curve (ESC) could be slightly expanded for clarity.

Response: An exclusive species exists solely in one habitat. The abundance matrix contains only one non-zero value (n_ij + 1), while all other values are (zero +1) to avoid the undefined calculation of log₂ (0). To construct the Exclusive Species Curve (ESC), we use Equation 3, which denotes Si in the calculation of the DP.

Comment 5: Justification of the original value (0.324) and its change to 0.3.

Response: In the normalized APD, the value 0.324 was obtained using an empirical method based on real data (see Dakki et al 2025), and we preferred the value 0.3 only by commodity and avoiding unnecessary decimal precision without altering the ecological meaning of the preference classes. [Page 5, lines 178–179]

Comment 6: The reason for deleting the 0.01 hyperbola.

Response: For low number of sampled habitats, the ESC shows an atypical shape at low abundances, and move away from the y-axis. [Page 3, lines 96–101 and Page 5, lines 180–182]

Comment 7: Add one or two sentences explaining the purpose of the optimization problem.

Response: This clarification is provided in [Pages 6–7, lines 196–199].

Comment 8: Why is DP < 0.1 problematic, and why were preference classes not shown in Figure 3?

Response: The preference class for accidental species was merged with the foreign species class. [Page 5, lines 182–183 and Page 6, lines 185–188]

Comment 8: The current “Results” section (Section 3) is confusingly structured. It starts by reminding the reader of the original APD method (3.1), then discusses the normalization (3.2), and finally mentions the algorithm (3.3). Sections 3.1 and 3.2 feel more like extended Methods or Introduction components.

Response: Thank you for this remark. To improve the manuscript's structure, we have moved the original content of sections 3.1 and 3.2 into a new dedicated Section 2: Reminder of the Abundance-Preference Diagram (APD) Method, placed immediately after the introduction (page 2). This section now consolidates the theoretical and methodological background

Comment 9: A new dedicated “Results” section should present the outcomes of applying the normalized method to the case studies, including new figures.

Response: A new paragraph has been included with new figures comparing the original and normalized versions of the APD to better illustrate the benefits of normalization in ecological interpretation. [page 7, lines 207–280]

Comment10: All figures require significant improvement. All figures need clear axis labels with units or descriptions.

Response: We have revised all figures (Figures 1–6) to improve visual clarity. Each figure now includes clearly labeled axes, with standardized terminology

Comment11: The Discussion section should interpret the results, highlight the advantages and limitations of the normalized method versus the original, discuss implications for community analysis, and suggest future directions. Case studies were mentioned but results were not compared to the original APD.

Response: We have thoroughly revised the Discussion section. We now emphasise how normalization overcomes the main limitation of the original method: the difficulty of comparing between assemblages with different demographic strategies has been solved using a diagram of square format. Additionally, we have incorporated the results of case studies (see Section 3.2) to demonstrate the practical application of the method to real ecological data, including comparisons between normalized and non-normalized APDs.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript addresses an important methodological issue by refining and normalizing the abundance-preference diagram (APD), clearly defines its objectives and demonstrates practical utility through a web-based application, thereby facilitating user engagement and accessibility for ecologists not specialized in statistics, while also showing strong potential for ecological applications such as analyzing species assemblages, community structure, and habitat preferences, all underpinned by a robust integration of statistical methods with ecological theory.

However, before accepting this manuscript I will need to have access to the web-based application (The link provided is not working) to test with my own datasets and verify its ecological relevance, for instance comparing communities following a geometric series, a log-series, a log-normal distribution or a a Brocken Stick series testing for the four main types of Species Abundace Distributions (SADs) in communities.

Indeed, the manuscript could strengthen its theoretical grounding by better contextualizing within current ecological literature on SADs. More recent references about community analysis, niche theory, and normalization techniques could enrich the paper.

Moreover, while the normalization method is mathematically sound, explicit justification of the choice for normalization (logarithmic transformation, base-2 log) relative to ecological interpretability could be expanded.

Some critical issues are also:

• Absence of explicit comparative evaluation against other existing community analysis methods (e.g., NMDS, PCA, or canonical correspondence analysis). Such comparisons would help validate the advantages claimed by the APD method
• As I mentioned above, the manuscript could significantly benefit from additional validation examples with broader ecological datasets (communities following a geometric series, a log-series, a log-normal distribution or a a Brocken Stick series) to ensure robustness and demonstrate versatility,

Author Response

We would like sincerely thank reviewers for their thoughtful and constructive feedback on our manuscript. Their insightful comments have been instrumental in improving the clarity, structure, and scientific rigor of our work. We have carefully revised the manuscript in accordance with their suggestions, including a more precise theoretical positioning, clearer methodological explanations, and the inclusion of applied results on real datasets.

Here are some key answers to your suggestions and questions

Comment 1: The access to the web application (the provided link does not work) to test it with my own datasets and verify its ecological relevance, e.g., by comparing communities following geometric, logarithmic, log-normal or broken-stick series—the four major Species Abundance Distribution (SAD) types.

Response: The web application designed to implement the APD method (both original and normalized) is part of a complementary research project. It is currently being finalized and will be submitted as a separate software article for publication. For this reason, a stable public link was not available at the time of this manuscript submission. It will be soon (6-7 days) available.

Comment 2: The choice of logarithmic transformation to base 2 for normalizing abundances in your APD method.

Response: The usefulness of the log₂ transformation is well established in both mathematics and ecology. In different methods implemented in community ecology, this transformation allows reducing the high influence played by very abundant species on rare species.

Comment 3: Absence of explicit comparative evaluation against other existing community analysis methods (e.g., NMDS, PCA, or canonical correspondence analysis). Such comparisons would help validate the advantages claimed by the APD method.

As I mentioned above, the manuscript could significantly benefit from additional validation examples with broader ecological datasets (communities following a geometric series, a log-series, a log-normal distribution or a a Brocken Stick series) to ensure robustness and demonstrate versatility,

Response: The primary objective of this study is to formalize and normalize the APD method while providing an accessible implementation tool. A comprehensive comparative evaluation with these classical methods is planned for a future dedicated study, which will focus on the complementarity and discriminative capacities of APD.

Nevertheless, preliminary tests using PCA and hierarchical clustering were conducted on the same datasets presented in this study. These methods proved to be less effective at distinguishing species based on their ecological specialization, particularly in the presence of species with very low or very high abundances. The signals of these species are often diluted or masked in standard multivariate projections.

We also agree with the suggestion to assess the robustness of APD using theoretical community structures (e.g. geometric series, log-series, log-normal distribution or broken stick models). While such analyses exceed the scope of the current manuscript, a second article is in preparation. This article is specifically dedicated to evaluating the APD method on simulated communities that follow these classical patterns of species abundance distributions. This complementary study will further validate the ecological robustness of APD.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I have read through the revised version of the manuscript now and it seem to clarify some previous structural questions.

I commend the authors for their efforts to enhance the overall clarity of the manuscript in this revised version. These improvements, particularly in the presentation of the methodology and the proposed algorithm, are constructive and contribute positively to the readability of the paper. However, I continue to have concerns regarding the theoretical grounding of the proposed approach and its relevance to a broader community of ecologists focused on biodiversity and species assemblage analysis.

More specifically, the ecological interpretation of the normalized classification of species based on hyperbolic classes remains insufficiently addressed. While this classification is more extensively discussed in a previous study by the authors (Dakki M., Ennakri M., El Farouki M., Ziti S. & Zoglat A. (2025). A Graphical Method for Analysing Species Assemblages Combining Habitat Preference and Abundance. African Journal of Ecology, 63(1), pp. 1–13), it would be beneficial to include a concise yet comprehensive explanation of the ecological meaning of each class within the present manuscript. Terms such as accidental, foreign, transgressive, preferential, elective, and characteristic should be clearly defined within an ecological framework.

For example, what does it mean to classify a species as transgressive or elective in ecological terms? The term transgressive species has a well-established meaning in the context of hybridization and niche divergence, where hybrid individuals exhibit phenotypes outside the range of the parent taxa (Rieseberg, L., Archer, M. & Wayne, R. Transgressive segregation, adaptation and speciation. Heredity 83, 363–372 (1999). https://doi.org/10.1038/sj.hdy.6886170). This classical interpretation appears to diverge from the intended use in the manuscript, and clarifying this distinction is important for readers from diverse ecological backgrounds.

Overall, I believe that the normalized ADP method presented in the manuscript offers a potentially valuable contribution to the toolbox of multivariate methods for classifying and quantifying species–habitat relationships (such as IndVal, OMI, and TITAN). Nevertheless, providing a more explicit definition of the ecological implications of the classification categories, and clarifying how these contribute to advancing our understanding of community structure and species–habitat associations, would enhance the manuscript’s impact and accessibility to a broader audience within the Diversity readership.

Major concern:

- This article addresses a relevant issue in community ecology: the need to standardize analytical methods to enable robust comparisons between different types of assemblages. The proposed standardization of the "Abundance-Preference Diagram" (APD) is a valuable contribution. However, the manuscript requires significant improvements in its English language, structuring of arguments to meet MDPI's publication standards, and contextualization with existing methods (e.g., IndVal, TITAN, OMI).

- In their conclusion, the authors state that they "...are improving the website interface, adding some relevant functions, expanding the range of acceptable data types for processing, creating a practical guide to facilitate its use and an aid for interpreting its results." I believe these improvements should appear in this manuscript as supplementary material or as part of a fully functional API. This would improve access to the proposed methods for readers and community ecology practitioners.

Minor suggestions:
- Add a simplified example figure of the APD with commented ecological interpretation.

- Figure 1: Add a more descriptive caption explaining what hyperbola curves are and how to interpret the zones.
- Provide a working link to the API (the link to https://apd.biodivrersity.ma/, dons't work in the time of my review!).
- Make sure that to meet english standards for consision and clarity. 

Comments on the Quality of English Language

The authors have made some efforts to address my previous suggestions. However, to meet the broader audience within the Diversity readership I suggest the the manuscript to be reviewd to meet language standards for consision and clarity. The text should be reviewed by a native English speaker or a professional editing service to correct multiple errors and improve fluency.

Author Response

We would like to thank the reviewer sincerely for their constructive and detailed feedback, which has helped us to significantly improve the quality and clarity of the manuscript.

Comment 1:
I commend the authors for their efforts to enhance the overall clarity of the manuscript in this revised version. These improvements, particularly in the presentation of the methodology and the proposed algorithm, are constructive and contribute positively to the readability of the paper. However, I continue to have concerns regarding the theoretical grounding of the proposed approach and its relevance to a broader community of ecologists focused on biodiversity and species assemblage analysis.

More specifically, the ecological interpretation of the normalized classification of species based on hyperbolic classes remains insufficiently addressed. While this classification is more extensively discussed in a previous study by the authors (Dakki M., Ennakri M., El Farouki M., Ziti S. & Zoglat A. (2025). A Graphical Method for Analysing Species Assemblages Combining Habitat Preference and Abundance. African Journal of Ecology, 63(1), pp. 1–13), it would be beneficial to include a concise yet comprehensive explanation of the ecological meaning of each class within the present manuscript. Terms such as accidental, foreign, transgressive, preferential, elective, and characteristic should be clearly defined within an ecological framework.

For example, what does it mean to classify a species as transgressive or elective in ecological terms? The term transgressive species has a well-established meaning in the context of hybridization and niche divergence, where hybrid individuals exhibit phenotypes outside the range of the parent taxa (Rieseberg, L., Archer, M. & Wayne, R. Transgressive segregation, adaptation and speciation. Heredity 83, 363–372 (1999). https://doi.org/10.1038/sj.hdy.6886170). This classical interpretation appears to diverge from the intended use in the manuscript, and clarifying this distinction is important for readers from diverse ecological backgrounds.

Overall, I believe that the normalized APD method presented in the manuscript offers a potentially valuable contribution to the toolbox of multivariate methods for classifying and quantifying species–habitat relationships (such as IndVal, OMI, and TITAN). Nevertheless, providing a more explicit definition of the ecological implications of the classification categories, and clarifying how these contribute to advancing our understanding of community structure and species–habitat associations, would enhance the manuscript’s impact and accessibility to a broader audience within the Diversity readership.

Response 1:
We have clarified the ecological meaning of all APD classes by adding a paragraph explaining their role in species–habitat analysis. The revised text now distinguishes between species that show a high, moderate, or no statistical preference for a given habitat (e.g., characteristic, elective, or preferential versus transgressive, foreign, or accidental). We have also explicitly differentiated the meaning of transgressive in our method from its classical use in evolutionary genetics. These clarifications aim to improve interpretability and align the classification with established practices in community ecology.
[Page 3, lines 95–107]

 

Comment 2:
Figure 1: Add a more descriptive caption explaining what hyperbola curves are and how to interpret the zones.

Response 2:
Figure 1 caption has been updated to include a more descriptive explanation of the hyperbolic curves and their ecological interpretation.
[Page 4, lines 115–117]

 

Comment 3:
Provide a working link to the API (the link to https://apd.biodivrersity.ma/ doesn’t work at the time of my review!)

Response 3:
We confirm that the correct and working link is: https://apd.biodiversity.ma

 

Comment 4:
The authors have made some efforts to address my previous suggestions. However, to meet the broader audience within the Diversity readership, I suggest the manuscript be reviewed to meet language standards for concision and clarity. The text should be reviewed by a native English speaker or a professional editing service to correct multiple errors and improve fluency.

Response 4:
We carefully re-read the entire manuscript and took your remark into full consideration. Several sections have been revised to improve clarity, sentence structure, and conciseness. A professional English editing review has also been conducted to ensure fluency and adherence to academic language standards.

 

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for providing detailed replies to the review comments. While acknowledging your efforts to improve the manuscript, I remain concerned about two critical aspects that were not satisfactorily addressed:

 

Availability and Accessibility of the Web Application:

The inability to access and test the web application significantly hampers my ability to thoroughly evaluate the practical applicability and ecological relevance of the APD method. The manuscript explicitly promotes this web application as a key strength, facilitating the use of the APD method by ecologists. Thus, the absence of an accessible, functional link at this stage is problematic. Even if planned for future publication, providing temporary but stable access or alternative means (e.g., supplementary software/code) during the review process would have considerably strengthened your manuscript.

 

Validation with Theoretical Data and Comparative Analysis:

I appreciate the intention to publish further studies validating APD against theoretical community structures (geometric series, log-series, log-normal distribution, and broken stick models) and comparing APD's effectiveness with traditional ecological methods (e.g., NMDS, PCA, CCA). However, the current manuscript lacks essential evidence supporting the claimed advantages and robustness of the APD method. Such validation is fundamental to assess the method’s ecological significance and practical utility comprehensively. The manuscript’s assertions about APD's superiority in distinguishing species based on ecological specialization must be clearly substantiated through rigorous empirical or theoretical evidence within the current paper.

 

Given these concerns, I strongly recommend addressing these points explicitly by:

 

Providing immediate, even temporary, access to the APD web application for review purposes, or alternatively, supplying the algorithm/software in an accessible format.

 

Including at least preliminary comparative analyses or theoretical validations within this manuscript to robustly support the claims regarding the method’s advantages and versatility.

 

These revisions are crucial to ensure the manuscript's clarity, reproducibility, and practical relevance, essential aspects of rigorous scientific communication.

Author Response

We thank the reviewer for their critical and constructive feedback, which we took seriously. 

Comment: While acknowledging the efforts made, the reviewer raises two major concerns: (1) the unavailability of the APD web application prevents any direct evaluation of the method’s practical use; and (2) the manuscript lacks empirical or theoretical validation to support the claimed advantages of the APD approach over classical methods.

Response:

1. Web application access: A stable and functional version of the APD web platform is now accessible at https://apd.biodiversity.ma. This version allows users to upload ecological data, visualize results interactively, and export outputs.

2. Validation and comparative evidence:  We have included a concrete example in the revised manuscript. The factorial correspondence analysis (FCA), applied to the dataset from Supplementary Material 2, shows that assemblages KH and TF appear very close in the F2×F3 projection (Figure 6), suggesting similarity. However, the APD method reveals different species profiles between these same assemblages, underlining a clear ecological distinction not captured by FCA. This example demonstrates the added value of combining APD with traditional approaches, and supports the method’s capacity to detect subtler patterns in species–habitat associations.
   [Lines 351–361 and 374–382]

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors Dear Authors,   Thank you for the opportunity to review your manuscript. You address an important topic by proposing a normalization for the "Abundance-Preference Diagram" (APD), a useful graphical tool for analyzing species assemblages. Your initiative to make the method more robust for comparisons across different taxonomic groups and to facilitate its implementation via an algorithm is commendable and has significant scientific merit.   The work is well-grounded, and the rationale for the normalization is clear and scientifically sound. My feedback is intended to help you further refine the manuscript's presentation to ensure its contribution is communicated with maximum clarity and impact. I am recommending "Accept with Minor Revisions" and my suggestions are focused on enhancing the manuscript's polish.   Key Points for Revision:

1. Language and Readability: While the manuscript is understandable, it would benefit from a final, thorough language review to polish the prose. Refining sentence structures and addressing minor grammatical errors will improve readability and ensure the text flows smoothly. This will help your valuable scientific contribution shine through more clearly.
2. Methodological Clarity: Your methodology is robust. To further enhance it, consider slightly elaborating on the "why" behind certain steps, not just the "what." For example, when introducing Formula (5), briefly explaining the intuition behind using a logarithmic transformation of 'relative abundance' could be helpful for readers less familiar with the technique. Similarly, in section 3.2.4, integrating the algorithm's description more fluidly into the main methods text could improve the narrative. What about including at least a minimal description of FCA analysis and a breif sentences clarifying the correspondece between FCA and ADP methods? This would be helpful to set up ADP in the context of multidimensional analyses in community ecology.
3. Structure and Flow: The overall structure is logical. A minor adjustment could further improve the flow. For instance, the justification for normalization (section 3.1.1) could be more tightly integrated with the Methods section, allowing the Results section to focus more exclusively on presenting the outcomes of your normalized method and case studies. This is a small suggestion to streamline the reader's experience.
4. Figures: The figures are informative. For the final version, please ensure they are rendered at the highest possible resolution and that all labels in the axes and data points (especially the species-identifying numbers) are clearly legible and not overlapping.

Conclusion: Your work represents a significant and practical contribution to ecological data analysis. These suggestions are minor and aimed at refining the manuscript's presentation. I am confident that with these small adjustments, your manuscript will be an excellent and impactful publication.

I hope my comments are helpful in finalizing your work.

Author Response

I would like to sincerely thank you for the time and effort you devoted to reviewing our manuscript. Your insightful comments and constructive suggestions have greatly contributed to enhancing the clarity, accuracy, and overall scientific quality of our work.

We carefully considered each of your observations and integrated the necessary changes to strengthen the manuscript. Your feedback has been invaluable in improving both the structure and the readability of the study.

 

Comment 1: Language and Readability: While the manuscript is understandable, it would benefit from a final, thorough language review to polish the prose. Refining sentence structures and addressing minor grammatical errors will improve readability and ensure the text flows smoothly. This will help your valuable scientific contribution shine through more clearly.

Response 1: Thank you for this valuable comment. We fully agree with your observation. Therefore, we have carefully reviewed the entire manuscript to refine sentence structures, correct minor grammatical errors, and improve the overall flow of the text.

 

Comment 2: Your methodology is robust. To further enhance it, consider slightly elaborating on the "why" behind certain steps, not just the "what." For example, when introducing Formula (5), briefly explaining the intuition behind using a logarithmic transformation of 'relative abundance' could be helpful for readers less familiar with the technique.

Response 2: Thank you for the suggestion. We have added an explanation (lines 165–168) to clarify that the base-2 logarithmic transformation is used to reduce skewness and stabilise variance. This limits the influence of highly abundant species and slightly highlights low-abundance species, thus improving interpretability.

 

Comment 3: In section 3.2.4, integrating the algorithm's description more fluidly into the main methods text could improve the narrative.

Response 3: In the revised version, we have integrated the description of the algorithm more seamlessly into Section 3.2.4 of the main methodological narrative, rather than presenting it as an isolated block. The additional text (lines 278–288) describes the sequence of operations.

Comment 4: About including at least, a minimal description of FCA analysis and a brief sentence clarifying the correspondence between FCA and ADP methods.

Response 4: In the revised manuscript, we have added a concise statement (lines 368–371) that positions the APD method in relation to multivariate analyses such as FCA and PCA.